Once again, politics rears its ugly head. On one side, the folks charged with discovering and delivering the innovations (namely the researchers). On the other side, the folks charged with investing the nation’s IT research capital (namely the legislators). InfoWorld has an interesting article from last week in which the government attempts to explain the new IT spending initiatives. The fact that DARPA funding is now being channeled in new directions compared to the past is disturbing to many in the academic community. But according to Representative Dana Rohrabacher of California

The question is whether we should channel the amount of money being spent on research into esoteric projects at the universities that may or may not ever come to fruition and help anybody

Evidently, the academians and researchers called upon by the administration for advisory/recommendation purposes either aren’t being heard or are being ignored. Let’s hope that Rep. Rohrabacher’s understanding and opinion of research aren’t shared by the majority of his colleagues. Most academic researchers do meaningful research and they take it seriously as well. Furthermore, basic research has served as the foundation for the multitude of technological accomplishments achieved in this country. Federal budget repositioning isn’t new and no one wants to needlessly throw money away, but new directives issued by the administration are worrisome. The archive for webcasts by the House Committee on Science can be found here http://www.house.gov/science/webcast/index.htm. Check out the Computing Research Policy Blog as well as this ACM blog, which has more information on congressional hearings.

Previous CTWatch blog entries about the IT funding issues are below:
Revitalize HPC, but do it frugally
Federal supercomputing funding: Is it a consensus problem?
Former SecDef on the “technology base”

Research sharing is perhaps taking on a new model in the Netherlands. Citing the increasing financial restrictions for acquiring academic research material, the Dutch DAREnet was launched over a year ago as a portal and repository for thousands of documents from 16 major academic institutions. According to The Register, DAREnet is no longer in “demonstration only” mode and is now providing researchers with free material that previously would have been available only through journal subscriptions and other commercial outlets.

It should come as no surprise that the global IT markets have an increasing Asian presence. It has long been held that educational systems in countries like China and Japan place a greater premium on math and science than the US. The accuracy of that might be debatable. What’s not debatable is the commitment by the Chinese government to reach new levels of international technological strength. An article at SiliconValley.com describes the effectiveness of some of that country’s policies toward improving its technological standing.

While the article asserts that some of China’s new technological prowess has been both unethically and illegally obtained, it also exposes a potential weakness of our own domestic IT policies - lack of unification. Several blog entries on this site have discussed some of the problems with US policies and the seeming lack of commitment by the US government to clearly understand the importance of adequate funding for all of the components of cyberinfrastructure.

Need a supercomputer but can’t afford a Cray or time on IBM’s behemoth Blue/Gene? According to an article at New Scientist, you might have an option besides stringing together all of the neighborhood Dells. It seems a company in California is marketing a small supercomputer with a price tag of $100,000. Certainly not pocket change, but with a claimed peak power of 230 Gflops and about the size of a refrigerator, just slide the armoire out of the way. Sporting 96 1.2 GHz processors, this mini comes out at $435/Gflop (using simplistic math of course). Granted it wouldn’t make the last rung of the latest Top500 Supercomputer ladder and you wouldn’t use it to run a large, complex CFD simulation, but 230 Gflops is more than capable of crunching some solid numbers.

Such a machine could very well find its niche in smaller markets and bring computing power to those who need it but couldn’t really afford it. With IDC reporting that the high performance computing market grew roughly 23% from 2003 to 2004, which includes statistics of systems valued at less than $50K and more than $1M, the demand is clearly there.

The latest issue of IBM’s Journal of Research and Development is devoted to Blue Gene/L. It includes an overview of the architecture, reports on various subsystems, and a rundown of the software environment. Not yet available for order! Operators are not standing by! Just look at it online, you goon.

I suppose announcements of hardware acquisitions are becoming humdrum, unless you’re talking about BlueGene-level performance. But this story published today in the University of Oklahoma student daily caught my eye. First of all, the 6.5-teraflop Dell system they’re installing is being nicknamed “Topdawg,” and that’s just a cool name. And second, the system is being installed at the OU Supercomputing Center for Education and Research (OSCER). Notice that “education” comes first in that name, and in fact:

Henry Neeman, the director of Supercomputing for IT, said OSCER is the only university program of its kind in the world. The program teaches supercomputing to a wide range of people. A workshop called “Supercomputing in Plain English” teaches OU students about supercomputing through simple methods.

“We’ve focused on people who don’t know much about supercomputing,” Neeman said.

“Supercomputing in Plain English” — what a great idea!

(BTW, there is an error in the article regarding the production of the Top500 list, which is not, to my knowledge, compiled at NCSA. I’ve emailed the paper’s editor to suggest a correction. Maybe the confusion arose because NCSA’s computing environment includes a couple of Dell systems — Tungsten and T2).